404 On definite Proportions. 



which combines with a greater portion, crystallizes, and 

 swims on the rest. If the operation of the column is 

 powerful, crystals are formed, which are small, irregular, 

 and sometimes needle-shaped, and commonly vegetate to- 

 wards the positive conductor, when hydrogen begins to be 

 evolved, and the decomposition of potass to be diminished, 

 if we neglect to push down the vegetation into the general 

 mass of the quicksilver. As the power of the column is 

 exhausted, the crystallization becomes more regular, and 

 sometimes very large hollow cubes are formed, consisting 

 of large quadrangular funnels, exactly like those of com- 

 mon salt. If these are collected, crushed into pieces, dried 

 on blotting paper, and exposed, in a closed vessel, to a tem- 

 perature of bO° Cels. [122®], they melt, and harden in 

 cooling into a crystalline crust, consisting of small solid 

 cubes, exactly as happens in the hasty evaporation of a 

 small portion of culinary salt. When treated with water, 

 their mass loses 'OlS/ of its weight, and consequently con- 

 tains little more than l| per cent, of potassium. 



If we distil an an)algiim of potassium in a small appa- 

 ratus filled with dried hydrogen gas, over the flame of a 

 spirit lamp, at first pure quicksilver passes over ; but after- 

 wards, when the metals remain in nearly equal volumes, 

 some potassium accompanies it ; and lastly, when nothing 

 more passes over at a low red heat, there is found left in 

 the retort a melted metallic substance, which when cold 

 adheres so firmly to the glass, that the retort must be 

 broken in order to obtain it. In the flexure of the retort 

 some congealed drops are always found, perfectly resembling 

 an amalgam of lead or tin. The residuum has a faint me- 

 tallic splendour, is of a gray colour, inclining to red, and 

 changing in a short time in the open air into dark brown 

 or black : it is by no means pure potassium, this substance 

 being, according to Davy, fluid at a moderate temperature, 

 like quicksilver, but greativ resembles the protoxide of 

 potassium, which Davy obtained by melting dry potass 

 with potassium. When thrown into water, this substance 

 sinks immediately to the bottom, and hydrogen gas is 

 evolved with the greatest violence; at last a globule of 

 quicksilver lemains, which occupies -^ oiilv of its original 

 volume. If this residuum is heated in the flame of a can- 

 dle, it swells and changes into a saline mass, but does not 

 inflame, I have proposed to myself three questions re- 

 specting this substance, neither of which I can satisfactorily 

 answer: Is it a combination of the protoxide of potassium 

 v/iili quicksilver, that is, of a metal with an oxidated sub- 

 stance ? 



